Research and Application of Machine Vision Algorithms for Defect Detection in Additive Technologies
Year: 2025
Type: article
Abstract: This paper discusses solving the problem of visualizing and recognizing defects that arise during selective laser melting using machine vision algorithms. The main goal of defect recognition is to reduce the time spent on selecting the technological parameters of the additive manufacturing equipment using automation methods of analysis of results. The paper presents an approach to visualizing and detecting defects that occur during the leveling stage of the metal powder layer. A methodology for software and hardware defect detection is considered and implemented. A general approach to image processing and analysis using a conveyor method is developed. The paper also discusses issues related to layer-by-layer photo documentation of the leveling process to simplify the analysis of the causes of defects. The developed software module can detect defects at the initial stages of production, allowing the process of printing knowingly defective products to be stopped, thereby enabling faster adjustment of the technological parameters of the equipment. This approach significantly reduces the time interval spent on selecting the technological parameters of the equipment and it reduces the cost of selective laser melting by saving metal powder on printing defective products. The advantages and disadvantages of the work done and the results obtained are presented.
Source: Scientific Visualization
DOI: 10.26583/sv.17.1.09
Authors: O.N. Tretyakova, D.N. Tuzhilin, A.A. Shamordin
Institution: Moscow Aviation Institute
Language: English
Subfield: Industrial and Manufacturing Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4409727135
